Soil Pollution Information Provision and Stakeholder Response

How to cite

Long, Xianling; He, Guojun. 2025. "Soil Pollution Information Provision and Stakeholder Response." AEA RCT Registry. AEARCTR-0016641. https://doi.org/10.1257/rct.16641-1.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Xianling Long* , Guojun He

* Primary investigator

Registry ID
AEARCTR-0016641
DOI
https://doi.org/10.1257/rct.16641-1.1
First registered
Last updated
Status
In development
Trial dates
2025-06-01 – 2027-01-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/16641

Abstract

This study investigates how information about soil pollution affects market behavior. We will conduct a randomized controlled trial (RCT) to examine how different forms of information disclosure (private vs public) affect decisions by intermediaries, sellers, and buyers. The experiment is designed to identify both direct effects of information on treated units and general equilibrium spillovers to untreated units. By varying the recipient of private information, we also test how asymmetric access to environmental information shapes strategic behavior. The results will contribute to our understanding of market responses to environmental information, the behavioral consequences of selective transparency, and the design of information disclosure policies.

Experimental design

We conduct a randomized controlled trial to study how information about local soil pollution affects market behavior on a digital platform. The units of randomization are neighborhoods within cities, and the intervention consists of disclosing environmental risk information in different formats and to different recipients.

The experimental arms include: a control group receiving general public information, private information disclosure to intermediaries, private information disclosure to sellers, and public disclosure.

The trial is cross-randomized at the city level to vary treatment intensity (high vs low share of treated neighborhoods) and the share of private vs public disclosure. The design allows us to estimate both the direct impact of information provision on treated units and indirect spillovers on untreated units.

Design details

Randomization method
Randomization done by a computer
Randomization unit
Three levels of randomization: city level, neighborhood level, and home level.
Clusters
more than 30,000 neighborhoods
Observations
more than 30,000 neighborhoods
Arms
more than 20,000 neighborhoods divided into 5 treatment arms
Primary outcomes
responses to our information; citizen attention; market equilibrium results.

Cited by

No citation data available for this registration yet.

Keywords